244 BOTANICAL GAZETTE [SEPTEMBER 
spores of Uromyces caryophyllinus from Euphorbia Gerardiana Jacq. collected | 
near Heidelberg infected Tunica prolifera (L.) Scop. and rarely Saponaria 
ocymoides L., while aecidiospores from the same host collected in the Wallis, 
Switzerland, infected Saponaria ocymoides. The relations of the last form to 
Tunica prolifera were not determined. Further cultures have now shown that 
the form from the Wallis infects both Saponaria ocymoides and Tunica prolifera 
and to some extent also T. Saxifraga (L.) Scop. Uredospores from Saponaria 
ocymoides or from Tunica prolifera infect either of those hosts indifferently. 
The cultures with Puccinia Pulsatillae Kalchb. from Anemone montana 
Atragene alpina L. Comparing the specialization of these forms with that of 
Uromyces corvopiathinier: fruidaree istinguishes two types. The first is cor- 
related with the geographical distribution of the hosts and is illustrated by 
Uromyces aiceeaiian In the Wallis, where both Saponaria and Tunica are 
common, this rust occurs on both of these plants, whereas in Baden, where 
prolifera to such an extent that it scarcely infects Saponaria. Specialized races . 
of this type show no distinctive morphological characteristics by which they 
might be separated from each other. The second type of specialization is 
correlated with the degree of affinity of the host plants. The forms of rusts 
showing this type of specialization each inhabit groups of closely related species 
of host plants and do not cross readily from one group to another. The races 
showing this type of specialization usually have slight morphological differ- 
ences, besides their biological behavior, by which they can be distinguished. 
Iro”? reports the successful infection of Pourthiaea villosa Decne. with 
teleutospores of Gymnosporangium Photiniae (P. Henn.) Kern (G. japonicum 
Syd.) from stems of Juniperus chinensis L. Pyrus sinensis Lindl., P. Malus L., 
and Amelanchier asiatica Koch were not infected. These cultures show that 
inhabiting form which the author identifies with Gymnosporangium Haraeanum 
Syd. (G. asiaticum Miyabe) which, according to the experiments of SHIRAI 
and those of Hara cited by the author, belongs to Roestelia koreaensis P. Henn. 
on leaves of Pyrus sinensis. SuHtrat does not state whether he used the leaf- 
inhabiting form or the stem-form in his experiments. 
sin following papers were Agere in 1914. 
RoMME™ successfully infected Myrica cerifera L. with teleutospores of 
aac Ellisii (Berk. ; Farlow from Chamaecyparis thyoides L. This 
Ero, S., — Notizen iiber parasitische Pilze Japans. Bot. Mag. Tokyo 
27:217-223. 19 
. ee F. D., A new gymnosporangial connection. Mycologia 6: 226-230. 
1914. : 
